Please benchmark my rehab costs per square foot across [PROJECT NAMES, or say: my last ten completed projects].
1. Pull each project record for the square footage, property type, year built and city. Skip any project with no square footage on file and tell me which ones you skipped, because they cannot be normalized.
2. Pull the expense line items for each project and total them by expense category. Use the totals the API returns for the full filtered set, not the totals of the page you happened to read.
3. Build a grid with expense categories down the side and projects across the top. In each cell give me the total spend and the cost per square foot.
4. Add a portfolio average column for each category and a total row at the bottom for overall cost per square foot per project.
5. Call out the outliers. Any category where a project is more than 40 percent above or below the portfolio average, with the dollar difference next to it.
6. Note anything that makes a comparison unfair. A project with large uncategorized spend, a project with very few line items, or houses different enough in age or type that comparing them per foot is misleading.
Change nothing.
Square footage comes from the project record, so the benchmark is only as good as that field. Projects with it blank get skipped and named rather than estimated.
Expense line items are read one project at a time. Ten projects is a comfortable run. Fifty is slow, so scope it to a comparable set rather than everything you have ever done.
This compares actual spend only. The API does not expose Rehab Estimator data, so estimate against actual is not possible here and is not implied anywhere in the output.
You have finished a dozen rehabs and you still cannot say what a kitchen costs you. Each project dashboard shows its own numbers and nothing puts them next to each other.
Total spend is not the answer either, because a 2,400 square foot house should cost more than a 1,100 square foot house. Without normalizing for size you cannot tell whether a trade is consistently overpriced or the house was just bigger, so you keep taking bids without knowing what good looks like.
Your assistant pulls the square footage off each project record and the expense line items behind it, then divides one by the other to give you cost per square foot by category. Categories run down the side, projects across the top, with a portfolio average in the last column.
Raw totals cannot answer whether a kitchen was expensive, because a bigger house costs more. Cost per foot can. That is what turns eleven finished projects into a pricing model you can hold a contractor to.
What it will not do: this automation is read only. It reports what you actually spent and changes nothing. It has no access to your estimates, so it cannot tell you whether you were over budget, only what the work cost per foot.
This prompt requires a connection to the FlipperForce Public API. Once your AI agent is connected to your account, you can run any prompt in the library.
Learn how to connect your AI agent to the FlipperForce Public API
Before you run this, check:
Best run on completed projects. A project halfway through construction looks cheap per foot because the work is not finished.